Original article
Cymiazole, a systemic acaricide that controls Acarapis
woodi (Rennie) infesting honey bees. II. An apiary test*
F. A. Eischen
D. Cardoso-Tamez
A. Dietz
1
G. O. Ware
1
Department of Entomology, University of Georgia, Athens, GA 30602, USA;
Apicultura Cardoso, Apdo. Postal No. 1, Buenavista Allende, Nuevo Leon, Mexico;
3
Agriculture Experiment Station, University of Georgia, Athens, GA 30602, USA
2
(received 1-12-1987, accepted 1-8-1988)
bee colonies located in northeastern Mexico were grouped according to the
prevalence of Acarapis woodi parasitizing their workers. Lightly (:
5 10%), moderately (20—60%), and
heavily (90—100%) infested colonies (N 30 each group) were randomly assigned to control or
treated subgroups (N 15). Controls received 50% sucrose syrup, and treated colonies received
syrup medicated with 0.3 mg/ml of cymiazole (2-[2,4-dimethylphenyl-imino]-3-methyl-4-thiazolinehydrochloride, Apitol
, GRA, 17.5% cymiazole, Ciba-Geigy Ltd.). Small, medium, and large colonies
s
received 300, 600, and 1 000 ml of syrup, respectively. Syrup was fed 3 times at weekly intervals.
At the end of treatment, the parasite load scores in medicated colonies were 0.53, 0.84, and 0.63
in the lightly, moderately, and heavily infested groups, respectively. This compared with 2.25, 3.43,
and 1.13 parasite load scores in the control colonies of the same groups, respectively (all P < 0.05).
In each group some colonies did not respond typically to medication. Mite populations at the end of
the test were positively associated with colony sizes and food reserves at the beginning of the test.
Increased dosages, as well as redistribution/reduction of food reserves are recommended for greater efficacy.
Unexpectedly, medicated colonies had significantly larger bee populations than controls at the
end of the test (P < 0.05). Increased bee longevity due to parasite load reduction is suggested as the
Summary &mdash; Honey
=
=
cause.
Acarapis woodl- cymiazole
-
dosage
Résumé &mdash; Le cymiazole (Apitol
), un acaricide systémique contre Acarapis woodi (Rennie),
&reg;
parasite de l’abeille. Il. Test au rucher. On a choisi 3 groupes de colonies dabeilles (N 30)
situées dans le nord-est du Mexique (lat. 25
1 12’N, long. 99° 58’ 0) en fonction de leur parasitisme
par Acarapis woodi : léger (5 10°l), moyen (20-60%), fort (90-100
Y.). Chaque groupe a été
,
ensuite subdivisé de façon aléatoire en sous-groupes traités et sous-groupes témoins (N
15). Les
témoins ont reçu un sirop de saccharose à 50% et les colonies traitées un sirop supplémenté avec
0,3 mglml de cymiazole (hydrochlorure de 2-(2,4-diméthylphényl-iminoj-3 méthyl-4-thiazoline; spécialité Apitol
O de Ciba-Geigy Ltd. sous forme de granulés dosés à 17,5% de cymiazole). Les petites
colonies, les moyennes et les grosses ont reçu respectivement 300, 600 et1 000 mi de sirop 3 fois
de suite à une semaine d’intervalle.
=
=
’
Mention of a commercial or proprietary product in this paper does not constitute an
by the University of Georgia or the Food and Drug Administration.
Present address : Department of Entomology, Pullman, WA 99164, USA
"
endorsement of the
product
A la fin du traitement le taux de parasitisme des colonies traitées était respectivement de 0,53,
0,84 et 0,63 dans les groupes légèrement, moyennement et fortement infestés, contre 2,25, 3,43 et
1,13 chez les témoins des mêmes groupes (P < 0,05). Dans chaque groupe quelques colonies
n’ont pas réagi au traitement de façon typique. Les populations d’acariens en fin de test étaient
positivement corrélées avec la taille de la colonie et les réserves de nourriture en début de test.
Pour une plus grande efficacité, on recommande des doses plus fortes ainsi qu’une redistribution
réduction des réserves de nourriture.
Contre toute attente, les colonies traitées avaient en fin de test une population significativement
plus forte (P < 0,05) que les témoins. On pense que la longévité accrue des abeilles dueà la diminution du parasitisme en est la cause.
ou une
Acarapis woodi
-
cymiazole
-
dosage
2usammenfassung - Cymiazol (Apitol
), ein sytemisches Akarizid zur Bekämpfung von
&reg;
Acarapis woodl (Rennie) bei Honigblenen. 11. Ein Freilandversuch. Drei Gruppen von Völkern
der Honigbiene (N
30) in Nordost-Mexiko (lat. 25° 12’N, long. 99° 58’W) wurden gemä&szlig; ihrem
Befall mit Acarapis woodi (Rennie) ausgewählt. Leicht (
S 10%), mittel (2-60
) und stark (90&mdash;
%
100%) befallene Völker wurden anschlie&szlig;end wahllos auf Kontroll- oder Versuchsuntergruppen
(N 15) verteilt. Die Kontrollgruppen erhielten 50°ligen Saccharosesirup, die Versuchsvölker Sirup
mit 0.3 mglmi Cymiazol (2-[2,4-Dimethylphenyl-iminoJ-3-methyl-4-thiazolin-hydrochlorid, Apitol
g
17.5% Cymiazol Granulat, Ciba-Geigy Ltd.). Die kleinen, mittleren und gro&szlig;en Völker erhielten drei=
=
mal 300, 600 oder 1 000 ml Sirup in wöchentlichem Abstand.
Nach der Behandlung war die Befallsrate der leicht, mittel und stark befallenen Völker 0.53, 0.84
und 0.64, im Vergleich zu einer Befallsrate von 2.25, 3.43 und 1.13 bei den Kontrollvölkern der gleichen Gruppe (P < 0.05). In jeder Gruppe gab es Völker, die nicht typisch auf die Medikation ansprachen. Die Milbenpopulation war am Ende des Versuchs positiv korreliert mit der Volksgrö&szlig;e und
den Futterreserven am Anfang des Tests. Um die Wirksamkeit des Mittels zu erhöhen wird empfohlen, stärker zu dosieren und die Futterreserven umzuverteilen bzw. zu reduzieren.
Entgegen den Erwartungen hatten die behandelten Völker im Vergleich zu den Kontrollvölkern
am Ende des Tests eine grö&szlig;ere Bienenpopulation (P < 0.05). Vermutlich ist die Ursache dafür in
einer längeren Lebensdauer aufgrund des reduzierten Befallsgrades zu suchen.
Acarapis woodi &mdash; Cymiazol &mdash; Dosierung
beekeeping in a difficult position. Safe
and efficacious control measures would
offer relief to infested areas and provide
assurance to uninfested regions that this
parasite is manageable.
Introduction
can
The rapid spread of the honey bee treacheal mite Acarapis woodi (Rennie)
throughout much of Mexico and the United States has complicated and frustrated
normal containment procedures (Guzman-Novoa and Zozaya-Rubio, 1984;
Anon., 1985; Anon., 1986b). The situation
has been exacerbated by its recent
discovery in Canada (Anon., 1986c;
Gruszka, 1987). A number of Canadian
provinces have responded by closing
their border to US shipments of queens
and package bees (Anon., 1986d). These
events left most sectors of North Ameri-
Recently the systemic acaricide cymiazole has been found efficacious in controlling the parasitic honey bee mite, Varroa
jacobsoni Oudemans (Ritter, 1985;
Schmid, 1985). Additionally, we have
found it effective in controlling A. woodi in
laboratory tests involving caged worker
bees (Eischen et al., 1989). This report
describes the performance of cymiazole
when fed to honey production colonies in
Mexico.
Materials and Methods
The following terms are used to describe infestations : prevalence(w), the percentage of
workers infested within a colony; parasite load,
the numbers of mites in an individual bee;
parasite load score, an estimate of the parasite
load. These terms are in keeping with current
parasitological terminology (Margolis et al.,
1982).
During early February 1987, 90 colonies of
honey bees located near Soto la Marina,
Tamaulipas, Mexico (lat. 23° 46’N, long. 98°
12’W) were selected for testing. Only colonies
with a laying queen, brood in all stages, and
bees across at least 3 frames were chosen.
Colonies were grouped according to the prevalence (w) of infested bees, viz., Group 1, < 10%,
Group II, 2D-60%, and Group 111, 90&mdash;100%
(N30). Infestation levels were determined by
removing 20 bees from the inner surface of the
hive cover, placing them in 80% ethyl alcohol
and dissecting them according to the technique
described by Eischen (1987). Bradbury (1924a,
b) has shown that a 20 bee sample gives reasonable accuracy except for the lowest infestations. Colonies were moved on 20 February
1987 to the test site near El Naranjo, Nuevo
Leon (lat. 25° 12’N, long. 99° 58’W). This location lies in the foothills of the Sierra Madre
Oriental and was chosen for its isolation from
other managed colonies and distance from
citrus orchards. The latter was important
because citrus trees in the general region were
expectad to flower in late March, and we
wished to ensure that the bees would not be
diverted from the medicated syrup.
Colonies within infestation groups (I, II and
III) were randomly assigned to either medicated
or unmedicated subgroups (N = 15). Colonies
in the 3 untreated subgroups were located ca.
100 m from the 3 treated subgroups. Distance
between subgroups ranged from 10&mdash;50 m.
Intercolony distance within subgroups was
2 m. These distances were chosen to reduce
intercolony drifting by workers.
=
A few days prior to treatment, colonies were
examined and estimates of the quantities of
bees, brood, honey, and pollen made in terms
of standard Langstroth frames (Kulincevic
et al., 1982). Additionally, a sample of 20 bees
was taken from each colony, placed in 80%
ethyl alcohol and subsequently an estimate of
their parasite load made, using the techniques
described by Eischen (1987).
Based on the number of adult bees, the 5
median colonies of each subgroup were selected for monitoring while the test was being
conducted. Ten bees from each of these colonies were removed weekly from the inner surface of the hive cover. Bees were dissected
and the number of living and dead mites counted according to the techniques described by
Eischen et al. (1987). Our purpose in monitoring these colonies was to determine how
selected dosages were affecting mite populations. This information gave us the option of
altering the dose if conditions warranted it.
However, no changes were made.
Colony treatment began on 28 February. All
medicated colonies received syrup containing
0.3 mg/ml of cymiazole (2-[2,4-dimethylphenyl-
imino]-3-methyl-4-thiazoline hydrochloride, Api17.5% cymiazole, Ciba-Geigy Ltd.).
One liter of medicated syrup was prepared by
dissolving 1.714 g of Apitol
&copy; in one liter of
warm (ca. 30°C) 50% sucrose syrup (vol/vol).
Enough medicated or unmedicated syrup
was prepared to treat one subgroup daily. The
quantity of syrup that each colony received was
based on the quantity of adult bees. Two small
colonies (< 3 frames covered with bees, both in
Group III, medicated) received 300 ml,
medium-sized colonies (3&mdash;7 frames) received
600 ml, and large colonies (> 7 frames) received 1000 ml. Syrup was placed in metal cans
(1600 ml) fitted with friction top lids in which
several holes had been punched. These cans
were inverted and placed over the top bars of
, GRA,
S
tot’
uppermost frames of the colony. Bees normally removed the syrup within 48 h. A few
the
colonies were slower, in which case after 72 h
the remaining syrup was sprinkled over the
bees. Colonies were fed 3 times at 7-day intervals. At the end of treatment, a light nectar flow
occurred. One standard shallow and one deep
super were given to all but the 2 extremely
small colonies in Group 111, medicated. Twentyfive days later, these supers were weighed.
The average of 5 supers of each size were
used to tare filled supers. Thirty days after the
last feeding, colonies were reexamined and
estimates of quantities of bees, brood, honey
and pollen made, along with an estimate of the
parasite load.
Colony estimates and mite data were evaluated with analysis of covariance (Snedecor
and Cochran, 1967). The objective was to
determine how the medication affected each
infestation group, i.e. Groups I, II, and 111. Mean
differences between beginning and ending
values within each group were tested for signifi-
after adjusting for initial starting treatsubgroup means. Additionally, infestation
percentages were given an arcsine transformacance
ment
tion.
Results
One colony died in each of the medicated
colonies of Groups 11 and 111. These
deaths were from queenlessness and
extreme dwindling, respectively, and were
not associated with the medication. Data
from these colonies have not been included in the analysis. Near the end of the
treatment period (15&mdash;25 March) several
colonies raised varying numbers of queen
cells, probably in preparation for swarming. None were found in the colonies of
Group 111. Approximately 50% of the medicated colonies in GroupsI and II raised
queens, while about 30% of the control
colonies of these groups did so. All colonies were examined weekly for queen
cells and when found were destroyed.
One swarm was found near the Group I
control colonies. Medicated colonies took
about 24 h longer to remove their syrup
than did controls. We have since learned
that Apitol
e may have reacted with the
metal containers. If so, palatability (but
not effectiveness) of treated syrups may
have been reduced (W.J. Schmid, pers.
comm.).
When
initially selected
on
7
February
1987, the percentages of bees parasitized
by A. woodi in Groups I, II and III averaged 6.0 ± 1.5, 42.4 ± 3.9, and 91.8 ±
1.6%, respectively. However, by 28
February when the test began, mite populations in the same groups averaged
21.0 ± 1.8, 47.3 ± 3.6, and 76.4 ± 3.9%,
respectively. That is, populations of A.
woodi had increased in Group I, remained
about the same in Group II, and had
decreased in Group III. This pattern of
change continued during the test period in
the untreated controls.
Mite population changes in the 5
median indicator colonies are shown in
Figure 1. In general, A. woodi populations
stabilized or declined in the medicated
colonies, while they rose in the untreated
colonies. The relative level of control
observed in the 5 colonies was indicative
of the entire subgroup (see TableI for
comparison). Seven days after the last
medication (28th day), infestations in the
5 control colonies of Groups I, II and III,
averaged 13.9, 15.6 and 10.5 mites/trachea, respectively. Medicated colonies of
the same groups at the end of the test
had on average 2.5, 8.9 and 5.1 mites in
their trachea, respectively.
A. woodi populations at the beginning
and end of the test are more completely
described by the data presented in
Table I. Parasite load scores show that
during the test, mite populations rose
255.7% in the control colonies of Group I.
Populations in the medicated colonies of
this group declined to 76.8% of their starting numbers (P < 0.01Group II data
were similar. Mite populations in control
colonies rose 137%, while medicated
colonies declined to 39.4% of their start-
ing population (P < 0.01in contrast,
parasite load scores declined in both the
control and medicated colonies of Group
III. The decline was greatest in the medicated colonies (P < 0.01Though parasite load scores cannot be converted to
precise mites/bee, the data indicate that
in the control colonies of Groups I, II and
III, there were roughly 22.5, 34.3 and 11.3
mites/bee, respectively at the end of the
test. Medicated colonies of the same
groups had roughly 5.3, 8.4 and 6.3
mites/bee, respectively.
Similar shifts in the mite populations
observed in the prevalence(w) of
infested bees. That is, in the untreated
colonies of Groups I and II, populations
rose to 173 and 143% respectively, while
populations fell in the medicated colonies
to 94 and 74% of their starting values,
respectively (both P < 0.01, Table I). The
number of infested bees in both the
control and medicated colonies dropped
in Group 111, with the decline being greatest in the medicated colonies (P < 0.01
).
The percentage of bees harboring bilateral infestations underwent changes similar to those described above. As with the
parasite load score and prevalence (w),
the number of bilateral infestions declined
in all the treated subgroups. Downward
shifts were most pronounced in colonies
bearing the heaviest infestations (Table I).
During the test, colony size, i.e. number of adult bees, increased in all treatment groups (Table II). Medicated colonies attained a greater size than controls
in all groups and were significantly larger
in Groups II (P < 0.05) and III (P< 0.01 ).
The percent increase in colony size was
were
greatest in Group 111. This may reflect the
increased reproductive efficiency of smaller colonies (Michener, 1964; Eischen
et al., 1983).
No significant differences were observed in brood production between control
and medicated colonies (Table II). The
quantity of honey in medicated and
control colonies did not differ significantly
at the start of the test or after the surplus
honey was harvested (Table II). Medicated colonies and greater pollen stores
than did controls. Differences were significant for GroupsI and II (both P < 0.05).
Honey production during the light nectar
flow by control colonies averaged 9.9, 8.7,
and 4.3 kg for Groups I, II and III, respectively. Medicated colonies of the same
groups produced 12.6, 7.7 and 5.2 kg,
respectively. These differences were not
significant. Though medicated colonies
had significantly more bees, more of them
attempted to swarm. This may have reduced honey production.
At the beginning of the test, parasite
load scores were negatively correlated
with the quantity of bees (r
-0.54,
P < 0.01brood (r
-0.48, P < 0.01),
honey (r -0.23, P < 0.05), and pollen
(r -0.29, P < 0.01; Table 111). Similar but
=
=
=
=
smaller correlation coefficients were found
for both the prevalence(w) of infested
bees, and the percentages of bees that
were bilaterally infested (Table 111). The
smaller r-values for these measures probably reflect a less accurate estimate of the
parasite loads.
In an attempt to identify factors affecting the performance of cymiazole, correlations were calculated between colony
populations at the start of the test with
those found 30 days after the last treatment. In control colonies, ending parasite
load scores in Group II were positively
correlated with the quantities of bees and
brood and at the beginning of the test
(P
<
0.05, Table IV). Similar correlations
found for medicated colonies in
Groups II and III (P < 0.05, Table V). Initial
quantities of honey and pollen were positively correlated with ending parasite load
scores in Group III (P <
0.05). A negative
correlation was found between the initial
and ending parasite loads of medicated
colonies (P < 0.01
).
were
Discussion
more
medicated syrup should have been
Additionally, colonies with larger
honey reserves at the beginning of the
test had higher ending mite populations.
Perhaps not needing the additional food,
the medicated syrup was only stored
(Moritz, 1982). This suggests that honey
fed.
may need to be redistributed or
reduced before treating. Should this
be imprudent, a topical application could
be employed (Anon., 1986a).
reserves
even
The rather dramatic mite
Medication with cymiazole resulted in
mite reduction. The level of
control achieved was about the same in
the 3 infestation groups. In each group
20% of the colonies did not respond typically to the medication. Why this occured
is not clear. A series of correlations was
made in an attempt to identify associated
factors. In general, larger colonies had
higher parasite levels, suggesting that
significant
=
population
changes in the control colonies affected
the efficacy estimates. Mite populations
expanded in both the lightly and moderately infested colonies, but dropped in the
heavily infested colonies. Eischen et al.
(1989) observed similar shifts during late
winter-early spring. The drop in heavily
infested colonies has been attributed to
the differential mortality of severely parasitized worker bees (Eischen et al., 1989).
A. woodi populations dropped in all medi-
cated colonies. Differences between treated and control colonies were largest in
the moderately infested colonies and
smallest in those that were heavily infested. That is, cymiazole was most effective
in moderately infested colonies and least
effective in heavily infested ones. Why
heavily infested colonies should be less
responsive is not clear. Perhaps differential mortality of heavily infested workers
simply obscured cymiazole’s potential efficacy. If this is true, it may be advantageous to medicate infested colonies in
autumn, thereby reducing spring mortality.
Unexpectedly, treatment with cymiazole was associated with significant
increases in bee populations. Ending
brood nest sizes
were
not
different, sug-
gesting that cymiazole promoted longevity
in bees. This could have been effected by
alleviating the stress of the parasite load
or by the lengthening life in a more direct
way. Eischen (1987) also observed that
infested colonies medicated with oxytetracycline HCI and fumagillin in autumn had
larger overwintering bee populations.
In summary, the level of control achienot complete, did reduce A.
woodi populations to at or below the suspected economic threshold, viz. 30% of
bees infested (Bailey, 1961; Eischen,
1987; Eischen et al., 1989). Based on
these observations and others (Dietz ett
al., unpublished) we believe colony
ved, while
could be raised 20&mdash;25%
without risk. This could probably be best
accomplished by raising the concentration
of cymiazole, along with somewhat greater quantities of syrup, especially for larger colonies. Application times and techniques that foster consumption of treated
syrups should also enhance efficacy.
manuki, USDA-ARS Beneficial Insects Laboratory, for advice and assistance while in Mexico.
Dorset Hurley provided technical assistance.
Acknowledgments
References
We wish to thank W.G. Hart, USDA-ARS Subtropical Insects Laboratory, W.T. Wilson,
USDA-ARS Honey Bee Laboratory, and H. Shi-
Anon.
dosages
We
are also grateful to the staff and families of
Apicultura Cardoso, who provided a great deal
of assistance as well as making our stay in
Mexico informative and memorable. Special
thanks are due to W.J. Schmid, Ciba-Geigy
Switzerland, who arranged for financial support
through Research Grant 25-21-RC293-108.
(1985) APHIS to propose federal mite
regulations be dropped. Speedy Bee 14 (9),
p. 2
Anon. (1986a) Apitol--readily available food for
bees. Bienenstich 8, p. 7
Anon. (1986b) Mite programs are under fire.
Speedy Bee 15 (1 ), 1-2
Anon. (1986c) Manitoba quarantines tracheal
mite infested colonies. Am. Bee. J. 126, 589
Anon. (1986d) East Canada closed to US bee
shipments. Speedy Bee 15 (3), p. 1
Bailey L. (1961) The natural incidence of Acarapis woodi (Rennie) and the winter mortality of
honey bee colonies. Bee World 42, 96-100
Bradbury F.C.S. (1924a) Some calculations on
acarine disease. Bee World 5, 145-147
Bradbury F.C.S. (1924b) Further calculations
on acarine disease. Bee World 6, 50-51
Eischen F.A. (1987) Overwintering performance of honey bee colonies heavily infested
with Acarapis woodi (Rennie). Apidologie 18,
293-304
Eischen F.A., Cardoso-Tamez D., Wilson W.T.
& Dietz A. (1989) Honey production of honey
bee colonies infested with Acarapis woodi
(Rennie). Apidologie 20, 1-8
Eischen F.A., Pettis J.S. & Dietz A. (1987) A
rapid method of evaluating compounds for the
control of Acarapis woodi (Rennie). Am. Bee J.
127, 99-101
Eischen F.A., Rothenbuhler W.C. & Kulincevic
J.M. (1983) Brood rearing associated with a
range of worker&mdash;larva ratios in the honey bee.
J. Apic. Res. 22, 163-168
Eischen F.A., Vergara C., Dietz A. & CardosoTamez D. (1988) Cymiazole, a systemic acaricide for the control of Acarapis woodi (Rennie)
infesting honey bees. I. Laboratory tests. Apidologie 19, 367-385
Gruszka J. (1987) Honey bee tracheal mites :
are they harmful ? Am. Bee J. 127, p. 653
Guzman-Novoa E. & Zozaya-Rubio A. (1984)
The effects of chemotherapy on the level of
infestation and production of honey in colonies
of honey bees with acariosis. Am. Bee J. 124,
669-672
Kulincevic J.M., Rothenbuhler W.C. & Rinderer
(1982) Disappearing disease. I. Effects of
certain protein sources given to honey-bee
colonies in Florida. Am. Bee J. 122, 189-191
Margolis L., Anderson R.C. and Holmes J.C.
(1982) Recommended usage of selected terms
in ecological and epidemiological parasitology.
4
Bull. Can. Soc. Zool. 13, p. 14
Michener C.D. (1964) Reproductive efficiency
in relation to colony size in hymenopterous
societies. Insectes Soc. 11, 317-341
T.E.
Moritz R.F.A. (1982) Praparatverteilung bei systemischer Therapie von Ektoparasitosen bei
Apis mellifera L. Apidologie 13, 127-141
Ritter W. (1985) First results from biological
trials with Apitol, a medicament with systemic
activity. Proc. 30th lnt. Apic. Congr., Nagoya,
Japan
Schmid W.J. (1985) Apitol
- a new acaricide
&reg;
with systemic activity against Varroa mites.
Proc. 30th lnt. Apic. Congr., Nagoya, Japan
Snedecor G.W. & Cochran W.G. (1967) Statistical Methods. The Iowa State University Press,
Ames, Iowa, pp. 593